Hacker News .hnnew | past | comments | ask | show | jobs | submit | hereme888's commentslogin

It's a big study about a very hot topic affecting real lives of people in every profession. Anti-trans or pro-trans, I believe it's important medical information for people to be as informed as possible for themselves and others.

Good for those who made the money. I would have done so had I had such insider info, as long as it didn't compromise any operation.

Do you not think that that would have a negative impact on the world? or do you not care, or is this "the world is going to hell so I may as well get mine"?

It's an unregulated world of gambling. Like trading DogeCoin. If you had insider knowledge, it was/is legal to trade there as you want.

Ah okay, so it's a case of "it's legal so it's fine", and impact or principles don't come into it.

Um, yes. If I were certain of a price movement before others, I'd take advantage, and so would you. But you can virtue signal with fake empathy if you'd like.

As long as the hordes of losers are not bothered by it enough to ask for fairness in the business, you’re right. Ethics and morals are constructs with the intention of creating a better, more just common existence. If the people in this common existence don’t care, then there is no reason for them not to do it.

> Opus 4.7 (Adaptive Reasoning, Max Effort) cost ~$4,406 to run the Artificial Analysis Intelligence Index, ~11% less than Opus 4.6 (Adaptive Reasoning, Max Effort, ~$4,970) despite scoring 4 points higher. This is driven by lower output token usage, even after accounting for Opus 4.7's new tokenizer. This metric does not account for cached input token discounts, which we will be incorporating into our cost calculations in the near future.

uhmmm so that is the p they are hacking? it would actually explain a lot

It's an incomplete cost model, but it's not p-hacking. Could be cherry-picking.

Do you still see Mythos as being so groundbreaking in cyber security now that it's widely reported previous and other models found the same vulnerabilities?

Yes, you're missing the forest for the trees. The impressive thing about Mythos is that it develops working exploits end to end without human guidance. Even Anthropic admitted 4.6 found most of the vulns. Finding a vuln is not that impressive, finding a working chain is the hard part.

OpenAI had been very strict about blocking reverse engineering/Ghidra/IDA_Pro-MCP tasks. I even got a warning email. I was having much more success convincing Claude Code for those tasks without warnings. Seems like they've tightened things up.

I switched to Codex. What I noticed is that while Claude was a more elegant coder and more accurate in how it went about coding, Codex is more intelligent... hard to describe it. If I could have a subscription to both, I'd use Opus to plan and code, and then check the work and fix issues with Codex.

Works with Oh-my-OpenCode? Does it change the terminal directory to the corresponding folder?

Another github repo to bypass another annoyance? They're so annoying.

don't think I need the MCP, but definitely downloaded the app itself. Seems great.

Thanks! We just released an update a few days ago with some new languages (Chinese/Turkish), and we now show idle time on the graph.

Didn't someone post recently that his team was unable to get OpenClaw to do anything productive ($) no matter the project?


Yup, that one.

Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: